Bias-Policy Iteration-Based Adaptive Dynamic Programming for Optimal Control of Discrete-Time Nonlinear Systems

This paper presents the bias-policy iteration, a modified adaptive dynamic programming method, to achieve optimal control design of discrete-time nonlinear systems. Firstly, the formulation of the bias-policy iteration method and the thorough convergence analysis are provided. By leveraging the bias...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on circuits and systems. I, Regular papers Jg. 72; H. 8; S. 4284 - 4296
Hauptverfasser: Jiang, Huaiyuan, Li, Xiang, Zhou, Bin, Cao, Xibin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.08.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Adaptive dynamic programming
Adaptive systems
Bias
bias-policy iteration
Circuits
Control systems
Convergence
Cost function
Discrete time systems
Dynamic programming
Iterative methods
Mathematical models
Nonlinear control
Nonlinear systems
Optimal control
Symmetric matrices
Vehicle dynamics
ISSN:1549-8328, 1558-0806
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper presents the bias-policy iteration, a modified adaptive dynamic programming method, to achieve optimal control design of discrete-time nonlinear systems. Firstly, the formulation of the bias-policy iteration method and the thorough convergence analysis are provided. By leveraging the bias parameter, the constraint of admissible control is relaxed while the fast convergence of traditional policy iteration is inherited. The actor-critic framework is utilized to realize the implementation of the proposed method accordingly. Finally, the proposed method is applied to optimal control problem of the inverted pendulum system. The simulation is conducted to verify the effectiveness of the bias-policy iteration approach.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1549-8328
1558-0806
DOI:10.1109/TCSI.2024.3492255